533d68380bbdc9076a147c0e1be4f50a4cff9118: Bug 861693 - Make DirectShow BaseFilter's destructor virtual, and move some code around to make our DirectShow BaseClass replacement easier to useoutside of webrtc module. r=jesup
Chris Pearce <cpearce@mozilla.com> - Tue, 13 Aug 2013 16:49:25 +1200 - rev 142371
Push 32377 by cpearce@mozilla.com at Tue, 13 Aug 2013 04:49:49 +0000
Bug 861693 - Make DirectShow BaseFilter's destructor virtual, and move some code around to make our DirectShow BaseClass replacement easier to useoutside of webrtc module. r=jesup
c4c323132512c267355bbb2acfe72acbde8d2ad1: Bug 861693 - Build system changes for DirectShow MP3 decoder HTML5 audio backend. r=ted
Chris Pearce <cpearce@mozilla.com> - Tue, 13 Aug 2013 16:49:25 +1200 - rev 142370
Push 32377 by cpearce@mozilla.com at Tue, 13 Aug 2013 04:49:49 +0000
Bug 861693 - Build system changes for DirectShow MP3 decoder HTML5 audio backend. r=ted
9cdd168a8518686f82cddde1e01218350d8b5d53: Bug 904108 - Add explicit assignment operator to js::Value r=luke
Adam Roach [:abr] <adam@nostrum.com> - Mon, 12 Aug 2013 12:01:23 -0500 - rev 142369
Push 32376 by adam@nostrum.com at Tue, 13 Aug 2013 03:41:06 +0000
Bug 904108 - Add explicit assignment operator to js::Value r=luke
a1d63e503ee91275d06431760e8fedfbe2af5900: Backed out changeset aadb205e7f05 (bug 881681) for Android Armv6 refteset-2 failures
Phil Ringnalda <philringnalda@gmail.com> - Mon, 12 Aug 2013 20:22:11 -0700 - rev 142368
Push 32375 by philringnalda@gmail.com at Tue, 13 Aug 2013 03:22:26 +0000
Backed out changeset aadb205e7f05 (bug 881681) for Android Armv6 refteset-2 failures
62ad090a94a4ae76ca6aaff167b15028be0384b3: b=815643 Implement HRTF panner processing based on Blink implementation r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Fri, 09 Aug 2013 10:08:06 +1200 - rev 142367
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=815643 Implement HRTF panner processing based on Blink implementation r=ehsan
746b2ba6cf30b83029ea892fa506c754f728913d: b=815643 Add Blink's HRTFPanner to the build r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Fri, 09 Aug 2013 10:07:49 +1200 - rev 142366
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=815643 Add Blink's HRTFPanner to the build r=ehsan
d92240f69c48e3db7d27e55d90e58d96a7fa2d72: b=815643 Refactor DelayNodeEngine delay processing into a shareable class r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Fri, 09 Aug 2013 10:07:49 +1200 - rev 142365
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=815643 Refactor DelayNodeEngine delay processing into a shareable class r=ehsan
7ab5c4babe56d9c00954d9f503b132afea499d26: b=865241 don't join HRTFDatabaseLoader thread until it has finished r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Fri, 09 Aug 2013 10:07:42 +1200 - rev 142364
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=865241 don't join HRTFDatabaseLoader thread until it has finished r=ehsan Also turn some always-true conditions into assertions.
c19e2f326d156dfa40341a34c3454004bff0d006: b=865241 make HRTFDatabaseLoader ref-counting thread-safe r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Fri, 09 Aug 2013 09:59:56 +1200 - rev 142363
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=865241 make HRTFDatabaseLoader ref-counting thread-safe r=ehsan
316df2d96b2420f285f9698cd2bacc819d4c9e5a: b=815643 Add Blink's HRTFDatabase and HRTFDatabaseLoader to the build r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Fri, 09 Aug 2013 09:56:28 +1200 - rev 142362
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=815643 Add Blink's HRTFDatabase and HRTFDatabaseLoader to the build r=ehsan
395fc5b9593cc2376f536012dc55e868c34164b5: b=815643 Add Blink's HRTFElevation to the build r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Thu, 08 Aug 2013 21:38:30 +1200 - rev 142361
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=815643 Add Blink's HRTFElevation to the build r=ehsan
5a05c22150916a815d67df4ec9850f1bba83c977: b=815643 Reduce convolver fft size for low sample rates r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Thu, 08 Aug 2013 21:38:30 +1200 - rev 142360
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=815643 Reduce convolver fft size for low sample rates r=ehsan
9b01d3f61a1bb50c16666203ff8c114b036f602a: b=815643 Add code to interpret and resample impulse response data r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Thu, 08 Aug 2013 21:38:30 +1200 - rev 142359
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=815643 Add code to interpret and resample impulse response data r=ehsan
c56ecdd125ecb0cb1f995786be8bfe934f775a48: b=815643 Add HRTF impulse response data r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Thu, 08 Aug 2013 21:38:30 +1200 - rev 142358
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=815643 Add HRTF impulse response data r=ehsan The impulse response sample data is directly from individual impulse response wave files in Blink SVN revision 153183.
c43863ae921862444a95cdaf271387c6923456c5: b=815643 Use symmetry to halve the number of HRTF kernels calculated and cached r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Thu, 08 Aug 2013 21:38:29 +1200 - rev 142357
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=815643 Use symmetry to halve the number of HRTF kernels calculated and cached r=ehsan
214033794f62382a72eacd6d58006c0313156cf0: b=815643 Add Blink's HRTFKernel to the build r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Thu, 08 Aug 2013 21:38:29 +1200 - rev 142356
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=815643 Add Blink's HRTFKernel to the build r=ehsan A functional difference is that HRTFKernel is not reference-counted.
5fd2ef2d994797f52b6cc510f077c7565ec540fa: b=865241 always leave 20 samples of headroom when extracting group delay, removing the discontinuity at delay = 20 r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Thu, 08 Aug 2013 21:38:29 +1200 - rev 142355
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=865241 always leave 20 samples of headroom when extracting group delay, removing the discontinuity at delay = 20 r=ehsan
729de18aacef823caea2635a9566eedc91f8c031: b=815643 Add Blink's frequency interpolation and group delay methods from FFTFrame to FFTBlock r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Thu, 08 Aug 2013 21:38:24 +1200 - rev 142354
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=815643 Add Blink's frequency interpolation and group delay methods from FFTFrame to FFTBlock r=ehsan
07e1d8379a30f1b3788625becaef972a9c623e59: b=865241 don't try to compute a phase for group delay from the DC and nyquist components of the DFT r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Thu, 08 Aug 2013 21:37:36 +1200 - rev 142353
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=865241 don't try to compute a phase for group delay from the DC and nyquist components of the DFT r=ehsan Given imagP[0] is really realP[halfSize], it doesn't make sense to calculate a phase from placing the DC and Nyquist components in different directions. The DC component doesn't have a phase offset and lastPhase is already initialized to zero. This makes extractAverageGroupDelay() consistent with addConstantGroupDelay() and to some extent interpolateFrequencyComponents(). The effect of the bug would have been small for reasonably sized FFTs.
5ddd533fc54cba690c22235c67a70b66b45d2bc5: b=815643 remove unused HRTF code r=ehsan
Karl Tomlinson <karlt+@karlt.net> - Thu, 08 Aug 2013 21:37:36 +1200 - rev 142352
Push 32374 by ktomlinson@mozilla.com at Tue, 13 Aug 2013 02:49:14 +0000
b=815643 remove unused HRTF code r=ehsan
(0) -100000 -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 +30000 +100000 +300000 tip